Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1762235AbXEUROZ (ORCPT ); Mon, 21 May 2007 13:14:25 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1756598AbXEUROR (ORCPT ); Mon, 21 May 2007 13:14:17 -0400 Received: from smtp23.orange.fr ([193.252.23.111]:8850 "EHLO smtp23.orange.fr" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755902AbXEUROQ (ORCPT ); Mon, 21 May 2007 13:14:16 -0400 X-Greylist: delayed 1391 seconds by postgrey-1.27 at vger.kernel.org; Mon, 21 May 2007 13:14:15 EDT X-ME-UUID: 20070521165103328.503CC700009D@mwinf2313.orange.fr Subject: Re: [RFC] enhancing the kernel's graphics subsystem From: Xavier Bestel To: Dave Airlie Cc: Jon Smirl , Jesse Barnes , Helge Hafting , Jesse Barnes , linux-kernel@vger.kernel.org In-Reply-To: <21d7e9970705210927s1454dc5ete89005715253f4b7@mail.gmail.com> References: <200705171423.46748.jesse.barnes@intel.com> <21d7e9970705210227x27234e2dme4e731a53f977082@mail.gmail.com> <46516A14.2000102@aitel.hist.no> <200705210857.44911.jbarnes@virtuousgeek.org> <9e4733910705210907g70142adfhcaf01bd6791f113d@mail.gmail.com> <21d7e9970705210927s1454dc5ete89005715253f4b7@mail.gmail.com> Content-Type: text/plain Date: Mon, 21 May 2007 18:50:55 +0200 Message-Id: <1179766255.7699.216.camel@frg-rhel40-em64t-04> Mime-Version: 1.0 X-Mailer: Evolution 2.0.2 (2.0.2-27) Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1134 Lines: 26 On Mon, 2007-05-21 at 17:27 +0100, Dave Airlie wrote: > > What about modifying the existing fbdev API? You could start with > one > > fbdev device per CRTC and then add a new IOCTL to control the output > > device. I haven't seen anything yet that justifies abandoning the > > existing fbdev API. > > Then you can't aribtrate properly output hooking is a root level > thing, you cannot allow the user in multiseat to just pick his own > outputs, if you claim to want a truly flexible interfaces, also the > crtc->output mappings aren't always simple, there are limitations on > most hw about which crtc can map to which output and when you can > clone etc.. putting policy for this stuff in-kernel would heavily > restrict what the user can do... Policy for this kind of thing should just go in HAL's ConsoleKit: http://lists.freedesktop.org/archives/hal/2007-January/007111.html Xav - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/